Node Classification On Pubmed 60 20 20 Random
المقاييس
1:1 Accuracy
النتائج
نتائج أداء النماذج المختلفة على هذا المعيار القياسي
جدول المقارنة
| اسم النموذج | 1:1 Accuracy |
|---|---|
| revisiting-heterophily-for-graph-neural | 90.81 ± 0.52 |
| simple-and-deep-graph-convolutional-networks-1 | 89.98 ± 0.52 |
| revisiting-heterophily-for-graph-neural | 90.09 ± 0.68 |
| revisiting-heterophily-for-graph-neural | 91.31 ± 0.6 |
| revisiting-heterophily-for-graph-neural | 87.75 ± 0.88 |
| mixhop-higher-order-graph-convolution | 87.04 ± 4.10 |
| neighborhood-homophily-guided-graph-1 | 91.56 ± 0.50 |
| revisiting-heterophily-for-graph-neural | 90.46 ± 0.69 |
| revisiting-heterophily-for-graph-neural | 90.39 ± 0.33 |
| joint-adaptive-feature-smoothing-and-topology | 85.07 ± 0.09 |
| revisiting-heterophily-for-graph-neural | 90.96 ± 0.62 |
| revisiting-heterophily-for-graph-neural | 90.18 ± 0.51 |
| simplifying-graph-convolutional-networks | 85.5 ± 0.76 |
| revisiting-heterophily-for-graph-neural | 90.63 ± 0.56 |
| predict-then-propagate-graph-neural-networks | 85.02 ± 0.09 |
| revisiting-heterophily-for-graph-neural | 90.56 ± 0.39 |
| generalizing-graph-neural-networks-beyond | 87.78 ± 0.28 |
| break-the-ceiling-stronger-multi-scale-deep | 88.8 ± 0.82 |
| revisiting-heterophily-for-graph-neural | 88.79 ± 0.5 |
| revisiting-heterophily-for-graph-neural | 86.43 ± 0.13 |
| gnndld-graph-neural-network-with-directional | 91.95±0.19 |
| revisiting-heterophily-for-graph-neural | 90.12 ± 0.4 |
| geom-gcn-geometric-graph-convolutional-1 | 90.05 |
| revisiting-heterophily-for-graph-neural | 89.15 ± 0.87 |
| inductive-representation-learning-on-large | 86.85 ± 0.11 |
| revisiting-heterophily-for-graph-neural | 90.74 ± 0.5 |
| simple-and-deep-graph-convolutional-networks-1 | 89.8 ± 0.3 |
| bernnet-learning-arbitrary-graph-spectral | 88.48 ± 0.41 |
| revisiting-heterophily-for-graph-neural | 91.44 ± 0.59 |
| simplifying-graph-convolutional-networks | 85.36 ± 0.52 |
| semi-supervised-classification-with-graph | 88.9 ± 0.32 |
| beyond-low-frequency-information-in-graph | 89.98 ± 0.54 |
| revisiting-heterophily-for-graph-neural | 90.66 ± 0.47 |
| the-split-matters-flat-minima-methods-for | 90.64 ± 0.46% |
| graph-attention-networks | 83.28 ± 0.12 |
| break-the-ceiling-stronger-multi-scale-deep | 89.04 ± 0.49 |
| node-oriented-spectral-filtering-for-graph | 89.89±0.68 |